Anda di halaman 1dari 4

Original article

Topographical distribution of blood supply to the anal canal


J. N. Lund, C. Binch*, J. McGrath, R. A. Sparrow* and J. H. Scholeeld
Departments of Surgery and *Anatomy, University Hospital, Nottingham NG7 2UH, UK Correspondence to: Mr J. N. Lund

Background: It has been suggested that anal ssure is an ischaemic ulcer caused by a combination of

poor blood supply to the posterior midline of the anal canal and spasm of the internal anal sphincter. This study investigated the topographical distribution of blood supply to quadrants of the anal canal above and below the dentate line. Methods: Cadaveric anal canals were removed and 1-cm blocks were cut above and below the dentate line. Blocks were sectioned at 10 lm and every 25th section was mounted. Using the technique of systematic random sampling, elds in the subanodermal space and the internal anal sphincter in posterior, lateral and anterior quadrants of the anal canal were chosen. The numbers of small arterioles in each eld were counted. Mean counts were compared for both subanodermal space and internal anal sphincter between quadrants and levels above and below the dentate line using Page's L test for trends. Results: Anal canals from eight cadavers were examined. There was a signicant trend to an increasing number of arterioles from posterior to anterior in the subanodermal space at all levels and at two of three levels in the internal anal sphincter. Conclusion: The arteriolar density is less in the posterior quadrant throughout the anal canal. It may be that this poor blood supply predisposes to the development of anal ssures at their most common site in the posterior midline.
Paper accepted 24 September 1998 British Journal of Surgery 1999, 86, 496498

Introduction

Materials and methods

Anal ssure is a split in the skin of the distal anal canal characterized by pain on defaecation and rectal bleeding. Recently it has been hypothesized that anal ssure may be an ischaemic ulcer caused by the combination of spasm of the internal anal sphincter (IAS) and poor blood supply to the posterior midline of the anal canal, the site at which the majority of ssures occur1,2. Only one study has previously examined the topography of the inferior rectal artery using post-mortem angiography1. Two distinct distributions of the blood supply to the anal canal were reported. In 15 per cent both the anterior and posterior midline of the anal canal were well supplied by branches of the inferior rectal artery. However, in the remainder (85 per cent) the posterior midline was poorly supplied. As anal ssure is a mucosal ulcer, the blood supply in the subanodermal space may also be decient relative to that in other quadrants of the anal canal. This study aimed to examine arteriolar density in the posterior, lateral and anterior quadrants of the subanodermal space and IAS above and below the dentate line.
British Journal of Surgery 1999, 86, 496498

Preparation of specimens The rectum and anal canal were removed en bloc from embalmed cadavers. The specimens were opened along the anterior midline and the dentate line was identied. Specimens were then transected at 1-cm intervals above and below the dentate line for the length of the anal canal. Blocks were xed in specipolymer wax and sliced on a microtome at 10 lm thickness. Every 25th section was mounted on a glass slide and stained with haemotoxylin and eosin. Arteriole counting and minimization of bias The technique of systematic random sampling was used to minimize bias3. One of every four slides was to be examined to ensure that arterioles in greater than ten sections per block were counted. Numbers from 1 to 4 were placed on paper and drawn from a `hat'. Every fourth slide from the initial starting point was then examined. Arterioles

496

www.bjs.co.uk

1999 Blackwell Science Ltd

J. N. Lund, C. Binch, J. McGrath, R. A. Sparrow and J. H. Scholeeld Blood supply to the anal canal 497

were identied as small vessels with a distinct thin muscular coat, making them easily distinguishable from venules or lymphatics. Arterioles were counted per graticule eld using the forbidden border technique3. On slides examined, a point corresponding to the posterior midline was positioned under the 200 objective of a microscope. The toss of a coin decided whether the eld was moved to the left or right by one eld of view. The same technique was used to select elds anteriorly or laterally. Numbers of arterioles per eld in posterior, lateral and anterior quadrants of the subanodermal space and the IAS were recorded on a spreadsheet (Excel 4.0; Microsoft, Seattle, Washington, USA). To estimate intraobserver variation the initial observer (J.N.L.) was blinded to the identity of the section and the number of vessels estimated in the same way as previously. Blinded counts were then compared with previous corresponding counts. To estimate interobserver variation a second independent observer (J.McG.) performed counts in the same way and the counts from each observer were compared. In each case ten sections were recounted, giving a total of 60 elds for comparison between and within observers (anterior, lateral and posterior for the subanodermal space and the IAS for each section). Analysis of results Comparisons between quadrants and levels were analysed using Page's L trend test for related samples4. This nonparametric test demonstrates whether or not the variable in question (in this case the number of arterioles per graticule eld) shows a signicant tendency to change in

the same direction across the samples in the order that the analysis is performed (in this case anterior to posterior). Statistical comparison of variation between and within observers was made using a two-tailed, paired t test.

Results

Eight cadaveric anal canals were examined, yielding a total of 960 mounted sections, 240 of which were analysed. The block of 2 cm above the dentate line for one subject was damaged in processing and does not appear in the results. No cadaver was known to have a history of anal ssure. Mean counts at each level are given in Table 1. The trend of arteriolar counts per eld was hypothesized to be posterior less than lateral less than anterior. Page's L trend values were calculated from mean values and are given with corresponding signicance levels in Table 2. There was a signicant trend to fewer arterioles posteriorly in the subanodermal space at all levels, and at 1 cm below and 2 cm above the dentate line in the IAS. There was no signicant trend when arteriolar densities between levels 1 and 2 cm above and 1 cm below the dentate line were compared. Using a two-tailed, paired t test to compare interobserver variation, there was no signicant difference between observers for counts in the subanodermal space (P 069) or in the IAS (P 029). Using the same test to compare intraobserver variation, there was no signicant difference between blinded counts for the subanodermal space (P 029) or for the IAS (P 021).

Table 1 Arteriole count per graticule eld ( 200 objective) for subanodermal space and internal anal sphincter
Submucosa Level 1 cm below dentate line (n = 8) 1 cm above dentate line (n = 8) 2 cm above dentate line (n = 7) Posterior 44(15) 35(10) 40(18) Lateral 61(17) 52(15) 53(11) Anterior 62(28) 52(14) 56(14) Internal anal sphincter Posterior 13(06) 14(07) 11(05) Lateral 21(11) 18(11) 17(06) Anterior 18(08) 17(09) 16(05)

Values are mean(s.d.) Table 2 Values of Page's L trend for the hypothesized trend towards decreasing number of arteriolar counts per graticule eld from anterior to posterior
Subanodermal space Level 1 cm below dentate line (n = 8) 1 cm above dentate line (n = 8) 2 cm above dentate line (n = 7) L 105 106 94 P < 005 < 001 < 001 Sphincter L 107 97 91 P < 005 > 005 < 005

If P < 005 the hypothesis is supported

1999 Blackwell Science Ltd

www.bjs.co.uk

British Journal of Surgery 1999, 86, 496498

498 Blood supply to the anal canal J. N. Lund, C. Binch, J. McGrath, R. A. Sparrow and J. H. Scholeeld

Discussion

This study demonstrates that there is a trend towards fewer arterioles in the posterior midline of the anal canal in the subanodermal space and the IAS compared with other quadrants. It is in the posterior midline that the vast majority of anal ssures occur. Although relatively few cadavers were analysed, the method yielded a great number of sections for analysis and the technique is reproducible and reliable3. Klosterhalfen et al.1 previously reported that there was a paucity of blood supply to the posterior midline of the anal canal in 85 per cent of cadavers examined using postmortem angiography. It has also been demonstrated by laser Doppler owmetry that blood ow in the posterior midline of the anal canal is less than that in other quadrants5. Using a different technique, the present work supports the hypothesis that the posterior midline of the anal canal is less well supplied with blood than other quadrants. The relative paucity of arterioles in the posterior quadrant of the anal canal was observed in all eight subjects, above and below the dentate line. Anal ssures occur below the dentate line and never extend more proximally and, although common, anal ssure does not affect everyone who may have this distribution of branches of the inferior rectal artery. The IAS encircles the anal canal above and below the dentate line but exerts its greatest pressure in the distal 12 cm of the anal canal6. The treatment of chronic anal ssure depends on returning the anal resting pressure to normal either by surgery or more recently by a `chemical sphincterotomy'. This return to normal pressure improves the blood supply to the posterior midline and results in the healing of the majority of

ssures7,8. It may be that the poor blood supply to the posterior midline of the anal canal is further compromised by the high resting pressure of the IAS almost invariably associated with posterior anal ssure, supporting the hypothesis that anal ssure is an ischaemic ulcer2. Approximately 10 per cent of anal ssures in females and 1 per cent in males are in the anterior midline9 and it may be that these lesions have a different aetiology.
References
1 Klosterhalfen B, Vogel P, Rixen H, Mittermayer C. Topography of the inferior rectal artery: a possible cause of chronic, primary anal ssure. Dis Colon Rectum 1989; 32: 4352. 2 Gibbons CP, Read NW. Anal hypertonia in ssures: cause or effect? Br J Surg 1986; 73: 4435. 3 Mayhew TM. The new stereological methods for interpreting functional morphology from slices of cells and organs. Exp Physiol 1991; 76: 63965. 4 Page EB. Ordered hypothesis for multiple treatments: a signicance test for linear ranks. J Am Stat Assoc 1963; 58: 21630. 5 Schouten WR, Briel JW, Auwerda JJ. Relationship between anal pressure and anodermal blood ow. The vascular pathogenesis of anal ssures. Dis Colon Rectum 1994; 37: 6649. 6 Keck JO, Staniunas RJ, Coller JA, Barrett RC, Oster ME. Computer-generated proles of the anal canal in patients with anal ssure. Dis Colon Rectum 1995; 38: 729. 7 Schouten WR, Briel JW, Auwerda JJA, De Graff EJR. Ischaemic nature of anal ssure. Br J Surg 1996; 83: 635. 8 Lund JN, Scholeeld JH. A randomised, prospective, doubleblind, placebo-controlled trial of glyceryl trinitrate ointment in the treatment of anal ssure. Lancet 1997; 349: 1114. 9 Keighley MRB. Fissure in ano. In: Keighley MRB, Williams NS, eds. Surgery of the Anus, Rectum and Colon. Vol. 1. London: WB Saunders, 1993: 36486.

British Journal of Surgery 1999, 86, 496498

www.bjs.co.uk

1999 Blackwell Science Ltd

Copyright of British Journal of Surgery is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.